CN107888981B

CN107888981B - Audio and video preloading method, device, equipment and storage medium

Info

Publication number: CN107888981B
Application number: CN201711139507.5A
Authority: CN
Inventors: 任金鹏
Original assignee: Beijing Xiaomi Mobile Software Co Ltd
Current assignee: Beijing Xiaomi Mobile Software Co Ltd
Priority date: 2017-11-16
Filing date: 2017-11-16
Publication date: 2020-12-18
Anticipated expiration: 2037-11-16
Also published as: CN107888981A

Abstract

The present disclosure provides an audio and video preloading method, apparatus, device and storage medium, wherein the method comprises: determining the pre-loading amount of the target audio/video based on the preference of a target user for the target audio/video; and preloading the target audio/video according to the determined preloading amount, wherein the audio/video data obtained by preloading is used for being played when receiving an instruction for playing the target audio/video. The embodiment of the disclosure not only pre-loads the audio/video, thereby increasing the play-starting speed of the audio/video, and realizing quick opening and playing of the audio/video; meanwhile, the pre-loading amount of the target audio/video is determined according to the preference of the target user to the target audio/video, so that the situation that all audio/videos are pre-loaded, excessive storage resources are occupied, and network flow is wasted is avoided.

Description

Audio and video preloading method, device, equipment and storage medium

Technical Field

The present application relates to the field of audio and video playing technologies, and in particular, to an audio and video preloading method, apparatus, device, and storage medium.

Background

The application and popularization of the internet bring great convenience to the life of users, and the users can carry out a series of activities such as work, study, entertainment and the like on the internet through the electronic equipment. When people watch audio/video through the internet, different audio/video files are generally downloaded and stored locally for watching. However, with the rapid increase of network speed, the audio-visual habit of people is shifted from traditional downloading and watching to online watching.

Different from the traditional audio/video playing mode of downloading and storing different audio/video files in the local for watching, the online playing is an audio/video playing mode which can transmit and play simultaneously so as to realize that a user can directly watch audio/video online. However, when the audio/video is loaded at the beginning of playing, the loading process is time-consuming due to the large amount of information loaded, thereby reducing the user experience.

Disclosure of Invention

To overcome the problems in the related art, the present disclosure provides an audio and video preloading method, apparatus, device, and storage medium.

According to a first aspect of embodiments of the present disclosure, there is provided an audio/video preloading method, the method including:

determining the pre-loading amount of the target audio/video based on the preference of a target user for the target audio/video;

and preloading the target audio/video according to the determined preloading amount, wherein the audio/video data obtained by preloading is used for being played when receiving an instruction for playing the target audio/video.

In an optional embodiment, the determining the pre-loading amount of the target audio/video based on the target audio/video preference of the target user includes:

predicting the click probability of the target user for clicking the target audio/video based on the user information of the target user, wherein the user information is used for determining the audio/video preference of the target user;

and determining the pre-loading amount of the target audio/video according to the click probability.

In an alternative embodiment, the user information includes one or more of a profile, an audio/video historical viewing record; the personal data of the target user comprises one or more of gender, age, occupation and location domain of the target user; the audio/video historical watching records comprise one or more of watched time length, click times, watching integrity and belonged classification of the audio/video.

In an optional embodiment, each item of information in the user information is used as a feature value of an interest feature, the user information corresponding to each audio/video forms a feature group, and the feature group comprises a feature value set of the interest feature; the method further comprises the following steps:

aiming at a feature group corresponding to audio/video, taking a feature value of an interest feature in the feature group as input data, and taking whether the video is clicked or not as output data to construct a training sample set;

training the initialization preference model by using the constructed training sample set to obtain the weight value of each interest feature;

the predicting of the click probability of the target user for clicking the target audio/video based on the user information of the target user comprises the following steps:

acquiring a feature group corresponding to a target audio/video;

and obtaining the click probability of the target audio/video clicked by the target user based on the obtained feature group and the weight value of the interest feature.

In an optional embodiment, the determining the pre-load amount of the target audio/video according to the click probability includes:

acquiring an influence factor of a conversion relation, wherein the conversion relation is the conversion relation between click probability and a pre-loading amount, the influence factor comprises one or more of a network type to which a current network belongs and available storage of a buffer area, and the network type comprises a WiFi network and a mobile data network;

and converting the click probability into the pre-loading amount of the target audio/video based on the conversion relation corresponding to the influence factor.

In an optional embodiment, the conversion relationship is a conversion ratio, the influence factor includes a network type to which a current network belongs, and the conversion ratio corresponding to the WiFi network is greater than the conversion ratio corresponding to the mobile data network;

or, the influence factor includes an available memory space of the buffer, and the conversion relation corresponding to the influence factor is: and the available memory space belongs to a conversion ratio corresponding to the capacity range, and the capacity range and the conversion ratio are in positive correlation.

determining the audio/video type of the target audio/video;

determining a pre-loading playing proportion corresponding to the determined audio/video type based on the corresponding relation between the audio/video type and the pre-loading playing proportion, and obtaining a pre-loading amount based on the pre-loading playing proportion;

wherein the pre-loaded playing proportion corresponding to the audio/video type in the corresponding relation is obtained based on a ratio, and the ratio is: the ratio of the viewed duration of the audio/video of said audio/video type viewed by the target user to the total duration of all audio/video viewed by the target user.

when the preloading condition is met, determining the preloading amount of the target audio/video based on the preference of the target user on the target audio/video;

the preload condition comprises one or more of the following conditions:

monitoring the starting of an application program for playing the audio/video;

monitoring a scroll stop event of an audio/video list;

the scroll stop event of the audio/video list is monitored and there is a target audio/video that is not preloaded.

In an alternative embodiment, the target audio/video is the audio/video currently shown in the screen in the audio/video list, or the target audio/video is the audio/video currently shown in the screen in the audio/video list and the audio/video to be shown in the screen after the scrolling operation is performed according to the page scrolling direction prediction.

In an optional embodiment, the method further comprises:

carrying out Hash operation on the URL address of the preloaded target audio/video to obtain a key code value;

establishing a corresponding relation between the key value and audio/video data obtained by preloading;

when a command for playing the target audio/video is received, indexing is carried out according to the key code value corresponding to the target audio/video so as to judge whether the target audio/video is preloaded or not;

and if the target audio/video is preloaded, playing the audio/video data obtained by preloading.

According to a second aspect of the embodiments of the present disclosure, there is provided an audio/video preloading device, the device comprising:

a pre-loading amount determining module configured to determine a pre-loading amount of a target audio/video based on a preference of a target user for the target audio/video;

and the preloading operation module is configured to preload the target audio/video according to the determined preloading amount, and the audio/video data obtained by preloading is used for being played when the target audio/video playing instruction is received.

In an optional embodiment, the pre-loading amount determination module includes:

a probability determination submodule configured to predict a click probability of a target user clicking a target audio/video based on user information of the target user, the user information being information for determining audio/video preference of the target user;

and the pre-loading amount determining sub-module is configured to determine the pre-loading amount of the target audio/video according to the click probability.

In an optional embodiment, each item of information in the user information is used as a feature value of an interest feature, the user information corresponding to each audio/video forms a feature group, and the feature group comprises a feature value set of the interest feature; the apparatus further comprises a weight value determination module configured to:

the probability determination submodule is specifically configured to:

acquiring a feature group corresponding to a target audio/video;

In an optional embodiment, the pre-loading amount determination sub-module includes:

the factor acquisition submodule is configured to acquire an influence factor of a conversion relation, wherein the conversion relation is the conversion relation between the click probability and the pre-loading capacity, the influence factor comprises one or more of a network type to which a current network belongs and the available storage capacity of a buffer area, and the network type comprises a WiFi network and a mobile data network;

and the information conversion submodule is configured to convert the click probability into the pre-loading amount of the target audio/video based on the conversion relation corresponding to the influence factor.

In an optional embodiment, the pre-loading amount determining module is specifically configured to:

determining the audio/video type of the target audio/video;

the preload condition comprises one or more of the following conditions:

monitoring the starting of an application program for playing the audio/video;

monitoring a scroll stop event of an audio/video list;

In an optional embodiment, the apparatus further comprises:

the hash operation module is configured to perform hash operation on the URL address of the preloaded target audio/video to obtain a key code value;

the relation establishing module is configured to establish a corresponding relation between the key value and the audio/video data obtained by preloading;

the information judgment module is configured to index according to a key code value corresponding to the target audio/video when the target audio/video playing instruction is received so as to judge whether the target audio/video is preloaded;

and the audio/video playing module is configured to play the audio/video data obtained by preloading if the target audio/video is preloaded.

According to a third aspect of the embodiments of the present disclosure, there is provided an electronic apparatus including:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to:

According to a fourth aspect of embodiments of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of any of the methods described above.

The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects:

the embodiment of the disclosure not only pre-loads the audio/video, thereby increasing the play-starting speed of the audio/video, and realizing quick opening and playing of the audio/video; meanwhile, the pre-loading amount of the target audio/video is determined according to the preference of the target user to the target audio/video, so that the situation that all audio/videos are pre-loaded, excessive storage resources are occupied, and network flow is wasted is avoided.

According to the method and the device, the preference degree of the target user to the target audio/video is represented by the click probability of the target audio/video clicked by the target user, so that different pre-loading amounts are corresponding to the target audio/video with different click probabilities, the pre-loading amount corresponding to the target audio/video with high click probability is large, and therefore when the target audio/video with high click probability is requested by the target user, the play starting speed is accelerated and the audio/video can be played more smoothly due to the large pre-loading amount.

According to the method and the device, the click probability of the target user clicking the target audio/video is predicted from a multi-dimensional angle, and the prediction accuracy can be improved.

The method comprises the steps of pre-constructing a preference model, and constructing a training sample set by taking the characteristic value of interest characteristics in a characteristic group as input data and taking whether a video is clicked or not as output data aiming at the characteristic group corresponding to the audio/video; the initialized preference model is trained by utilizing the constructed training sample set to obtain the weight value of each interest characteristic, and the influence degree of the interest characteristic on the audio/video preference of the target user can be reflected by the weight value of the interest characteristic, so that the click probability of the target audio/video clicked by the target user can be obtained on the basis of the characteristic group corresponding to the target audio/video and the weight value of the interest characteristic, and the efficiency and the accuracy of obtaining the click probability are improved.

According to the embodiment of the invention, the click probability is used as a factor for determining the pre-loading amount, and the network type of the current network is used as a factor for determining the pre-loading amount, so that excessive network flow consumption and loss of users caused by the fact that the large pre-loading amount is loaded under the mobile data network are avoided.

According to the embodiment of the disclosure, the click probability is used as a factor for determining the pre-loading amount, and the available storage amount of the buffer area is used as a factor for determining the pre-loading amount, so that the situation that when the available storage amount is small, the conversion proportion is still adopted when the available storage amount is large, and the pre-loading amount obtained based on conversion occupies a large storage space and possibly causes insufficient storage space is avoided.

In the embodiment of the disclosure, the preference of the target user for the target audio/video is determined by the watched time length of each type of audio/video watched by the target user, different pre-loading amounts are configured for different types of audio/video, and the method is easy to implement.

According to the embodiment of the disclosure, when the preloading condition is met, the preloading amount of the target audio/video is determined based on the preference of the target user to the target audio/video, so that resource waste caused by real-time calculation can be avoided.

The audio/video currently displayed in the screen is used as the coarse-grained screening condition in the embodiment of the disclosure, so that the audio/video currently displayed in the screen in the audio/video list is used as the target audio/video, and therefore, when the target user clicks the audio/video in the current page, the audio/video can be quickly opened and played, the processing task amount can be reduced, and the situation that resources are always occupied is avoided.

In the embodiment of the disclosure, not only the audio/video currently displayed in the screen in the audio/video list is used as the target audio/video, but also the audio/video to be displayed in the screen after the scrolling operation is performed according to the page scrolling direction prediction is used as the target audio/video, so that the audio/video possibly displayed on the screen is predicted in advance, the preloading operation is performed on the audio/video in advance, and the situation that the audio/video is played by fast clicking and the preloading operation is not performed yet is avoided.

According to the method and the device for pre-loading the audio/video data, the corresponding relation between the key code value and the audio/video data obtained through pre-loading is established, the audio/video data obtained through pre-loading by using the key code value can be pre-loaded, and the indexing efficiency can be improved.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and together with the description, serve to explain the principles of the disclosure.

Fig. 1 is a scene diagram illustrating an online playback/video according to an exemplary embodiment of the present disclosure.

Fig. 2 is a flow chart illustrating an audio/video preloading method according to an exemplary embodiment of the present disclosure.

FIG. 3 is a flow chart illustrating a method of preference model pre-construction according to an exemplary embodiment of the present disclosure.

Fig. 4 is a block diagram of an audio/video preloading device illustrated in accordance with an exemplary embodiment of the present disclosure.

Fig. 5-8 are block diagrams of another audio/video preloading device illustrated in accordance with an exemplary embodiment of the present disclosure.

Fig. 9 is a block diagram illustrating an apparatus for audio/video preloading according to an exemplary embodiment of the present disclosure.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the exemplary embodiments below are not intended to represent all implementations consistent with the present disclosure. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure, as detailed in the appended claims.

The terminology used in the present disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used in this disclosure and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, such information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present disclosure. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.

On-line audio/video playing is an important use scenario of electronic devices at present. As shown in fig. 1, fig. 1 is a scene diagram illustrating an online playback/video according to an exemplary embodiment of the present disclosure. The electronic device can play audio or video in the server through the network. The electronic device may be an electronic device with a playback/video function, such as a smart phone, a tablet computer, a Personal Digital Assistant (PDA), a television, and a multimedia player. In online audio/video playing, the play-starting speed of audio/video is an important index for measuring user experience. The playing starting speed of the online video can be the time from the time of clicking the playing video to the time of displaying the first frame of the video; the playing speed of the online audio can be the time from the audio playing by clicking to the audio playing of the first frame. The difference between online playing and local playing is that the storage positions of audio/video data are different, so that when clicking playing, the audio/video data needs to be downloaded to a local buffer area. At present, when being clicked, the audio/video often needs to go through a loading process of establishing connection based on the address of the audio/video and performing buffer buffering, and the process is time-consuming and influences the play-up speed of the audio/video.

In order to improve the play-starting speed, the present disclosure provides an audio/video preloading method, which not only preloads audio/video, thereby improving the play-starting speed of audio/video, and realizing fast opening and playing audio/video; meanwhile, the pre-loading amount of the target audio/video is determined according to the preference of the target user to the target audio/video, so that the situation that all audio/videos are pre-loaded, excessive storage resources are occupied, and network flow is wasted is avoided.

The disclosed embodiments are described below with reference to the accompanying drawings.

As shown in fig. 2, fig. 2 is a flowchart illustrating an audio/video preloading method according to an exemplary embodiment of the present disclosure, which may be used in an electronic device having an audio/video playing function, and may include steps 202 and 204:

in step 202, determining the pre-loading amount of the target audio/video based on the preference of the target user for the target audio/video;

in step 204, the target audio/video is preloaded according to the determined amount of preloading, and the audio/video data obtained by preloading is used for being played when an instruction for playing the target audio/video is received.

In the embodiment of the present disclosure, the electronic device may be an electronic device with an audio/video playing function, such as a smart phone, a tablet computer, a PDA (Personal Digital Assistant), a television, and a multimedia player. An audio/video playing application program can be installed in the electronic equipment, and the audio/video playing application program can be utilized to play audio/video. In one example, a proxy mechanism may be added to an audio/video playback application for performing the disclosed method.

The audio/video playing application program can display the audio/video triggering control, the audio/video classification control and the like which can be clicked by the target user through the audio/video list. The target user may select an audio/video of interest from the audio/video list for playback. The audio/video trigger control can be presented in the form of text or images. The target user may be a logged-on user of the audio/video playback application.

The target audio/video in this embodiment may be an audio/video in an audio/video list.

In an alternative implementation, any audio/video in the audio/video list may be used as the target audio/video, so that all the audio/video in the list are preloaded as shown in fig. 2. However, in the case of a large number of audio/videos in the audio/video list, because there are often a large number of audio/videos in the audio/video list, and the target user generally only views a very small part of the audio/videos, the pre-loading operation as shown in fig. 2 is performed on all the audio/videos, which may cause a defect of resource occupation due to a large amount of processing tasks.

In one example, the target audio/video may be an audio/video currently shown on the screen in the audio/video list. Wherein the screen is a screen of the electronic device executing the method of the present embodiment. The audio/video currently presented in the screen may be the audio/video presented in the audio/video list window at the current time, and may also be referred to as the audio/video of the current screen. Since the page for displaying audio/video often has more page content, only part of the page content can be displayed in one window, and therefore, other part of the page content in the page can be viewed by sliding the scroll control or sliding the page. With respect to how to determine the audio/video currently displayed on the screen, the audio/video to be displayed on the screen may be determined according to the scrolling event, and other determination manners may also be adopted, which is not limited herein.

Because the audio/video currently displayed in the screen is the audio/video being viewed by the target user and the probability of being selected by the target user is higher, the audio/video currently displayed in the screen is taken as a coarse-grained screening condition to realize that the audio/video currently displayed in the screen in the audio/video list is taken as the target audio/video, so that the audio/video can be quickly opened and played when the target user clicks the audio/video in the current page, the processing task amount can be reduced, and the resources are prevented from being occupied all the time.

In practice, when the target user slides to the current screen, the user may quickly click on the playing audio/video, and there may be a situation where the time interval between the sliding to the current screen and the clicking of the audio/video is shorter than the time taken for pre-buffering, therefore, in order to further increase the play-out speed of the audio/video, in another example, not only the audio/video currently presented in the screen in the audio/video list but also the audio/video to be presented in the screen after performing the scroll operation is predicted according to the page scroll direction is taken as the target audio/video, thereby achieving early prediction of the audio/video that may be presented on the screen, the pre-loading operation is executed in advance for the audio/video, so that the situation that the audio/video is played by fast clicking and the pre-loading operation is not executed in time yet is avoided.

In this embodiment, the page scrolling direction may be a direction of the page content displayed after scrolling relative to the page content displayed before scrolling during the latest scrolling event from the current time, which may be understood as a moving direction of the page. After determining the page scroll direction, it is possible to predict audio/video that may be shown in the screen if a scroll operation is performed in the page scroll direction. As one of the prediction methods, since the layout of the page does not change in a short period of time, the page content adjacent to the page content currently displayed on the screen can be determined from the page along the page scrolling direction. The adjacent page content can be the page content which can be displayed by just one screen, or the page content with other specified sizes, and can be flexibly set.

The disclosed embodiments list several target audio/video, and it is understood that other coarse-grained screening conditions may be adopted to obtain the target audio/video.

As for the amount of pre-loading, the size of the pre-loaded obtained audio/video data of the target audio/video may be, for example, 2M. The amount of preload may also be a parameter that can characterize the size of the audio/video data obtained by preloading, for example, the length of preloaded playing time may be preloaded, or the preloaded playing proportion may be preloaded. The pre-loading playing time length is the time length that the pre-loading obtained audio/video can be played, and the pre-loading playing proportion can be the proportion that the pre-loading obtained audio/video accounts for the target audio/video.

The embodiment of the disclosure not only needs to load the target audio/video, but also needs to determine the pre-loading amount of the target audio/video, and the pre-loading amount is determined based on the preference of the target user, and the pre-loading operation is not performed on the target audio/video which is not preferred by the target user, and the pre-loading is performed only on the target audio/video which is preferred by the target user, so that the pre-loading judgment on the target audio/video is realized through the preference, and the defects of flow waste and occupation of more storage space caused by pre-loading all the target audio/video are avoided.

In an optional implementation manner, a trigger condition may also be set to step 202, so that the method of the present embodiment is executed only when the trigger condition is satisfied. For example, the amount of pre-loading of the target audio/video may be determined based on the target user's preference for the target audio/video when the pre-loading condition is satisfied. Several preloading conditions are listed below for illustration. The preload condition may include one or more of the following conditions:

the first condition is as follows: and monitoring the starting of an application program for playing the audio/video.

The video playing application program can be video software such as millet video, Tencent video, love art video and the like, and the audio playing application program can be audio software such as cool dog, Himalayan and the like. When an application starts, it is the first page of the application that is opened. Page contents such as audio/video trigger controls, audio/video classification controls and the like which can be clicked by a target user are often displayed on the home page. In the embodiment of the present disclosure, the audio/video in the top page may be used as the target audio/video, the pre-loading amount of the target audio/video may be determined according to the preference of the target user for the target audio/video, and the target audio/video may be pre-loaded according to the pre-loading amount.

And a second condition: the scroll stop event of the audio/video list is monitored.

Since the page content of a page often exceeds what can be presented by an application window, a scroll event may be set to enable the page content of the non-presented portion of the page to be presented by sliding the page or scroll bar. When the scroll stop event of the audio/video list is monitored, the target user can be presumed to be interested in the content of the current display page, and the interested audio/video can be clicked from the current display page for playing. In view of this, the amount of pre-loading of the target audio/video may be determined based on the target audio/video preference of the target user when the scroll stop event of the audio/video list is monitored. The target audio/video may be an audio/video currently shown in the screen in the audio/video list, or the target audio/video may be an audio/video currently shown in the screen in the audio/video list and an audio/video to be shown in the screen after performing the scroll operation according to the page scroll direction prediction.

And (3) carrying out a third condition: the scroll stop event of the audio/video list is monitored and there is a target audio/video that is not preloaded.

Wherein the non-preloaded target audio/video may be an audio/video for which the operations of

steps

202 and 204 are not performed. When the scroll stop event of the audio/video list is monitored, whether the target audio/video has been subjected to the operations of

steps

202 and 204 or not can be judged, if all the target audio/video has been subjected to the operations of

steps

202 and 204, the operations of

steps

202 and 204 are not carried out, and if the target audio/video does not have been subjected to the operations of

steps

202 and 204, the operations of

steps

202 and 204 are carried out on the target audio/video.

In one example, after the target audio/video is preloaded, hash operation may be performed on a URL (uniform resource locator) address of the preloaded target audio/video to obtain a key value; after the audio/video data obtained by preloading is played, the key value is emptied. Accordingly, when the scroll stop event of the audio/video list is monitored, whether the operations of

steps

202 and 204 are performed on the target audio/video is judged by whether the key value corresponding to the target audio/video exists.

For example, in the historical behavior, the operations in

steps

202 and 204 are already performed on the audio/video displayed in the first screen, and when the application program is opened this time and enters the first screen, if the audio/video displayed in the first screen is not changed, the operations in

steps

202 and 204 may not be performed. In practical applications, the content of the page may be updated, and the audio/video displayed in the first screen may be different from the audio/video displayed in the previous several views, so that in the first screen, part of the audio/video may have performed the operations of

steps

202 and 204, and part of the audio/video may not have performed the operations of

steps

202 and 204, and the audio/video that has not performed the operations of

steps

202 and 204 is processed, thereby avoiding resource waste caused by repeated processing.

Next, an example of how to determine the amount of preloading of the target audio/video according to the target user's preference for the target audio/video will be described.

In an alternative implementation, the user's preference for the target audio/video may be measured by using a specific value, so as to implement different pre-loading amounts for the target audio/video with different preference degrees.

In one example, the click probability of clicking the target audio/video by the user can be used for representing the preference degree of the user on the target audio/video, the click probability and the pre-loading amount can be in a positive correlation relationship, the larger the click probability is, the larger the pre-loading amount can be, the smaller the click probability is, the smaller the pre-loading amount is, and when the click probability is smaller than the lower limit value of the designated probability, the pre-loading amount can be zero. It will be appreciated that other parameters may be used to indicate the user's preference for the target audio/video, for example, the user's score for the target audio/video may be used to indicate the user's preference for the target audio/video, etc.

Specifically, the click probability of the target user clicking the target audio/video may be predicted based on the user information of the target user. Wherein the user information is information for determining audio/video preferences of a target user. The audio/video referred to herein is not specific audio/video. Information that can predict the audio/video preference of the target user can be taken as the user information referred to in the present embodiment. For example, the user information includes one or more of a profile, a historical viewing record of audio/video; and determining the pre-loading amount of the target audio/video according to the click probability.

The profile may be information describing a basic situation of the user. For example, the target user's profile may include one or more of the target user's gender, age, occupation, location domain. The audio/video history viewing record may be a record generated by the user viewing the audio/video. For example, the audio/video historical viewing record may include one or more of a viewed duration, a number of clicks, a viewing completeness, a category of the audio/video. In the embodiment of the disclosure, each item of information in the user information is taken as a feature value of an interest feature, that is, an attribute in the user information can be taken as an interest feature, and a content corresponding to the attribute in the user information can be taken as a feature value of the interest feature, so that the click probability of a target user clicking a target audio/video is predicted from the perspective of the multidimensional interest feature.

Because the audio/video favorite by the target users in different age stages, different sexes, different professions and different regions may be different, each attribute in the personal data can be used as an interest characteristic, and the interest characteristic is assigned according to the personal data of the target users to obtain a characteristic value. It is understood that the characteristic value referred to in the present embodiment is not limited to a specific numerical value, and may be other characters than a numerical value. For example, the personal data may include attributes such as gender, age, occupation, location, etc., and the personal data of the target user includes specific content corresponding to the various attributes. Therefore, the gender can be used as a one-dimensional interest characteristic, and the gender of the target user can be used as a characteristic value of the one-dimensional interest characteristic; the age may be used as a one-dimensional interest feature, and the age of the target user may be used as a feature value of the one-dimensional interest feature.

Since the history of watching audio/video by the target user can also reflect whether the target user prefers audio/video, each item of information in the history of watching audio/video can be used as the characteristic value of the interest characteristic. For example, the audio/video historical viewing record may include feature values of one-dimensional or multi-dimensional interest features in viewed duration, number of clicks, viewing completeness, belonging category, and the like of the audio/video.

In which a single file and multiple files may exist for the same audio/video, whether audio or video. Taking a video as an example, when the type of the video is a movie, the video is a single file; when the type of the video is a television play, the video is a plurality of files, and each set can be used as an independent video file. Therefore, when the video is a single file, the watched duration may be the playing duration of the watched portion of the video file; when the video is a plurality of files, the viewed duration may be the sum of the play durations of the viewed portions of all the video files.

Regarding the number of clicks, the number of clicks of the audio/video by the target user may be used. In this embodiment, the degree of preference of the target user for such audio/video may be reflected by the number of clicks.

Regarding viewing completeness, it may be a proportion of a viewed part to a complete part for a certain audio/video. Since the target user tends to view the favorite audio/video in its entirety, the viewing integrity is also one of the factors reflecting the degree of the preference of the target user for the audio/video.

As for the belonging classification, classification from different dimensions, coarse classification and fine classification can be performed for audio/video. For example, videos may be roughly classified into movies, dramas, fantasy, news, games, animations, short videos, etc., and the videos may be finely classified into coarse categories and into different dimensions, for example, movies may be finely classified into dimensions of movie origins, inland, hong kong, usa, japan, etc.; the method can be divided into fine categories such as action, love, comedy, thriller and science magic according to the dimension of the story; the method can be divided into subclasses such as 2017, 2016 and 2015 according to the time dimension; the scores may be divided by the dimension of the score into 5 or less, 5 to 6, 6 to 7, 7 to 8, 8 to 9, 9 to 10. As another example, audio may be broadly classified into voiced books, children, vocal commentary, music, talk shows, and the like. The vocal book can be classified into suspense, city, literature, martial arts and the like.

It will be appreciated that depending on the dimensions of the division, introduction of audio/video from different dimensions may be implemented. The number of categories to which audio/video belongs may be one category or a plurality of categories without limitation. For example, for video a, its belonging categories may include: movies, inland, action movies, and the like.

Further, the interest feature is not limited to include the above-listed interest features, and may include other interest features as long as it can be a factor for judging whether the target user likes audio/video. For example, the interest feature can also be a label of audio/video, and the label is also an introduction of the audio/video to make up for the limitation of fixed classification, and the dimension of introducing the audio/video can be expanded. For example, the tags may be affective tags, which are used to rate audio/video. For example, the audio/video belongs to emotional categories of joy, sadness, fear, and the like. The label can be obtained by extracting keywords from the name of the audio/video, can be a label created by an administrator for the audio/video, and can also be a label created by a user for the audio/video, so as to introduce the audio/video from the perspective of the user. For example, the user may label the audio/video after viewing it, or vote for an existing label. In order to avoid the situation that the same audio/video has too many tags, the tags with high occurrence frequency or high vote number can be screened according to the occurrence frequency of the tags or the vote number of the tags, and the screened tags are used as the tags of the audio/video.

Since the audio/video historical viewing records correspond to the audio/video, and the personal data can also be used as a factor for judging whether the target user likes the audio/video, the user information can comprise the personal data and the historical viewing records of the audio/video for each audio/video corresponding to the user information. Therefore, the preference relationship between the target user and the audio/video can be embodied by using the historical watching record of the audio/video, and the preference relationship between the target user and the audio/video can also be embodied by using the personal data of the target user.

Further, the user information corresponding to each audio/video may form a feature group, and the feature group includes a feature value set of the interest feature. The personal data comprises: sex is female, and age group is 20-30; the historical viewing record of the video comprises: the watched time is 60min, the number of clicks is 2, the watching integrity is 50%, and the video belongs to the categories of movies and suspensory films as examples, and then the interest characteristics of the video include: gender, age bracket, length of time viewed, number of clicks, completeness of view, category of affiliation, etc. The feature groups may be as follows:

sex woman

Age group 20-30

Viewed for 60min

Number of clicks 2

Viewing integrity 50%

Classified film and suspicion film

In order to predict the click probability of the target user for clicking the target audio/video, a preference model which can reflect the preference of the target user can be established in advance based on user information, and when a training sample is used for training, the weight value of each interest characteristic can be obtained, so that the preference model is obtained. For example, as shown in fig. 3, fig. 3 is a flowchart illustrating a method for pre-constructing a preference model according to an exemplary embodiment of the present disclosure, and the pre-creating step of the preference model includes:

in step 302, for a feature group corresponding to audio/video, a training sample set is constructed by taking a feature value of an interest feature in the feature group as input data and taking whether a video is clicked or not as output data;

in step 304, the initialized preference model is trained by using the constructed training sample set, and a weight value of each interest feature is obtained, so as to obtain the preference model.

The preference model obtained by training can calculate the click probability of the target user for clicking the audio/video based on the weight value of the interest characteristics, and determine whether the target user likes the audio/video based on the click probability. The weight value of an interest feature may reflect the degree of influence of the interest feature on the audio/video preferences of the target user. Therefore, the method and the device for determining the pre-loading amount of the target audio/video can calculate the click probability of the target audio/video clicked by the target user by using the obtained weighted value of the interest characteristics, and further determine the pre-loading amount of the target audio/video according to the click probability.

The input data in the training sample can be the characteristic value of a single interest characteristic, or can be a key value pair consisting of the characteristic values of two interest characteristics in a characteristic group, and the input data can be input data by combining the key value pair as the training sample, so that the dimensionality of the input data can be widened, and a preference model with higher judgment accuracy can be obtained by training.

After the weighted value of the interest feature is obtained, when online judgment is performed, a feature group corresponding to the target audio/video can be obtained. When the target audio/video is the audio/video which has been watched by the user, the feature value of each interest feature in the feature group corresponding to the target audio/video is not null, and when the target audio/video is the audio/video which has not been watched by the user, the feature value of a part of interest features in the feature group corresponding to the target audio/video is null.

As one implementation manner, the empty feature value may be set to 0, and then the click probability of the target user clicking the target audio/video is obtained based on the obtained feature group and the weight value of the interest feature. For example, the feature value of the interest feature and the weight value of the interest feature may be subjected to weighted summation to obtain the click probability of the target user clicking the target audio/video; or the characteristic values in the characteristic group corresponding to the target audio/video can be input into the preference model, and the click probability of the target user for clicking the target audio/video can be determined by using the preference model.

As another implementation manner, aiming at the condition that the feature value of part of interest features in the feature group corresponding to the target audio/video is empty, other users having similarity with the target user are obtained, the feature value of the interest features is determined according to the historical watching records of the target audio/video watched by the other users, and the feature value is used as the feature value of the interest features in the feature group corresponding to the target audio/video, so that the interest features in the feature group are assigned. After the assignment is successful, the click probability of the target user clicking the target audio/video can be obtained based on the obtained feature group and the weight value of the interest feature. For example, the feature value of the interest feature and the weight value of the interest feature may be subjected to weighted summation to obtain the click probability of the target user clicking the target audio/video; or the characteristic values in the characteristic group corresponding to the target audio/video can be input into the preference model, and the click probability of the target user for clicking the target audio/video can be determined by using the preference model.

The above embodiments only list several ways for determining the click probability, and other ways may also be adopted to obtain the click probability of the target user clicking the target audio/video, which are not listed here.

After the click probability is obtained, the pre-loading amount of the target audio/video can be determined according to the click probability. Wherein, there may be a conversion relationship between the click probability and the pre-load amount.

In one example, the transition relationship may be a correspondence, e.g., different click probability ranges correspond to different amounts of pre-load.

In another example, the conversion relationship may be a proportional relationship, for example, a conversion ratio is preset, and after the click probability is determined, the click probability is multiplied by the conversion ratio, so as to obtain the pre-load amount. Further, the target audio/video with the click probability being too small is not preloaded, so that the pre-loading amount of the target audio/video corresponding to the click probability being less than the lower limit of the probability is set to be 0.

Further, in order to avoid resource waste caused by excessive preloading amount, an upper limit value of the preloading amount may be preset, and when the determined preloading amount is greater than the upper limit value of the preloading amount, the preloading amount may be directly updated to the upper limit value of the preloading amount. The amount of preload may be set flexibly, for example, according to the storage space of the electronic device.

In another example, the conversion relationship between the click probability and the pre-loading amount is different under different influence factors, so that the adaptive change is realized according to the influence factors. Specifically, the determining the pre-loading amount of the target audio/video according to the click probability may include:

acquiring an influence factor of a conversion relation, wherein the conversion relation is the conversion relation between click probability and a pre-loading amount, the influence factor comprises one or more of a network type to which a current network belongs and available storage of a buffer area, and the network type comprises a WiFi network and a mobile data network; and determining the pre-loading amount of the target audio/video according to the click probability and the conversion relation corresponding to the influence factor.

The conversion relation is the conversion relation between the click probability and the pre-loading amount, and different influence factors correspond to different conversion relations.

Under different types of networks, different conversion relations are configured for different types of networks based on consideration of factors such as network speed or network traffic, so that the click probability is used as a factor for determining the pre-loading amount, the network type of the current network is used as a factor for determining the pre-loading amount, and the conditions that the network speed influences pre-loading or causes consumption of a large amount of network traffic and the like are avoided.

In an example, the conversion relationship may be a conversion ratio, the influence factor may be a network type to which the current network belongs, the network type includes a WiFi network and a mobile data network, and the conversion ratio corresponding to the WiFi network is greater than the conversion ratio corresponding to the mobile data network, so that for the same click probability, a pre-loading amount obtained when the current network belongs to the WiFi network is greater than a pre-loading amount obtained when the current network belongs to the mobile data network. In view of this, determining the pre-load amount of the target audio/video according to the click probability may include:

determining the network type of a current network, wherein the network type comprises a WiFi network and a mobile data network;

acquiring a conversion ratio of the click probability corresponding to the determined network type and a pre-loading amount, wherein the conversion ratio corresponding to the WiFi network is larger than that corresponding to the mobile data network;

and determining the pre-loading amount of the target audio/video according to the click probability and the conversion ratio.

Wherein, the click probability and the conversion ratio can be multiplied to obtain the pre-loading amount. Therefore, in the embodiment, the click probability is used as a factor for determining the pre-loading amount, and the network type of the current network is also used as a factor for determining the pre-loading amount, so that the phenomenon that the large pre-loading amount is loaded under a mobile data network, which causes excessive network traffic consumption and loss to users is avoided.

In addition, the available storage capacity of the buffer area also influences the pre-loading capacity, so that the situation that the storage space of the buffer area is insufficient due to an excessively large pre-loading capacity is avoided, and the situation that the broadcast cannot be started quickly due to an excessively small pre-loading capacity is also avoided. In view of this, the conversion relation may be a conversion ratio, the impact factor may be an available storage amount of the buffer, and the conversion relation corresponding to the impact factor is: and the available memory space belongs to a conversion ratio corresponding to the capacity range, and the capacity range and the conversion ratio are in positive correlation. In view of this, determining the pre-load amount of the target audio/video according to the click probability may include:

determining the available memory space of a buffer area;

according to the capacity range to which the available memory belongs, acquiring the conversion ratio of the click probability and the pre-loading amount corresponding to the capacity range to which the available memory belongs, wherein the capacity range and the conversion ratio are in positive correlation;

Therefore, in the embodiment, the click probability is used as a factor for determining the pre-loading amount, and the available storage amount of the buffer area is also used as a factor for determining the pre-loading amount, so that the situation that when the available storage amount is small, the conversion proportion is still adopted when the available storage amount is large, and the pre-loading amount obtained based on the conversion occupies a large storage space and possibly causes insufficient storage space is avoided.

Furthermore, the type of the network to which the current network belongs and the available storage capacity of the buffer area can be simultaneously used as the influence factors of the conversion relationship, so that the common constraint conversion relationship is realized, and the situations that the storage space is insufficient and the mobile network flow is excessively consumed due to the overlarge pre-loading amount are avoided. It is understood that other influencing factors influencing the transformation relationship may also be included, which are not illustrated here.

In another alternative implementation, the target user's preference for the target audio/video may be reflected in preference results from two dimensions, e.g., preference (interest) and non-preference (non-interest). For both results, the amount of pre-loading of the non-preferred target audio/video is zero and the amount of pre-loading of the preferred target audio/video is a specified value.

In one example, a preference model for classification may be established in advance based on user information of a target user. The preference model functions as a model for predicting whether a target user prefers (is interested in) audio/video. In order to improve the prediction accuracy, the personalized preference model can be constructed through multiple dimensions. In view of this, each item of information in the user information may be taken as a dimension, that is, each item of attribute in the user information is taken as a one-dimensional interest feature, and the content corresponding to the attribute in the user information is taken as a feature value of the interest feature. The user information may include one or more of a profile, an audio/video historical viewing record, and the target user profile may include a target user's gender, age, occupation, and feature values of one-dimensional or multi-dimensional interest features in a locale. The audio/video historical watching records can comprise watched time length, click times, watching integrity and characteristic values of one-dimensional or multi-dimensional interest characteristics in the classification of the audio/video. For example, the pre-creation of the preference model includes:

and training the initialization preference model by using the constructed training sample set to obtain the preference model.

In the subsequent use process, the feature group corresponding to the target audio/video can be obtained, the feature values in the feature group are used as input data to be input into the preference model, the result of whether the target user likes the target audio/video is obtained, and the pre-loading amount is determined according to the result.

In another alternative implementation, the viewing duration of the target audio/video may be used to measure the target user's preference for the target audio/video. The viewing duration of the target audio/video is the duration that the target user is likely to view the target audio/video.

In one example, the viewing duration of the target audio/video may be predicted according to the audio/video preference of the target user, and the pre-load amount of the target audio/video may be determined based on the viewing duration. For example, the determining the pre-loading amount of the target audio/video based on the target audio/video preference of the target user may include:

predicting the watching time length of the target audio/video according to the type of the target audio/video and a preset time length model;

and determining the pre-loading amount of the target audio/video based on the conversion relation between the watching time length and the pre-loading amount and the watching time length obtained through prediction.

Wherein the preset duration model is used for predicting the watching duration of each type of audio/video. The preset duration model can be obtained by training according to the duration consumed by watching different types of audio/video in the audio/video historical watching records of the target user.

Therefore, the preference of the target user for the target audio/video is determined in a mode of predicting the watching time length of the target audio/video, so that the pre-loading amount is obtained.

In another alternative implementation, the determining the pre-loading amount of the target audio/video based on the target audio/video preference of the target user may include:

determining the audio/video type of the target audio/video;

The pre-loading playing proportion can be directly a ratio, that is, the pre-loading playing proportion corresponding to the audio/video type in the corresponding relationship is as follows: the ratio of the viewed duration of the audio/video of said audio/video type viewed by the target user to the total duration of all audio/video viewed by the target user. As other implementations, the pre-loaded play ratio value may also be affected by other factors, such as one or more of the type of the current network, including a WiFi network and a mobile data network, and the available storage capacity of the buffer. Then, the pre-loaded playback scale may be obtained based on the ratio and the impact factor.

Furthermore, the watched time lengths of the audio/video of various audio/video types can be predicted according to various audio/video types and a preset time length model, after the watched time lengths of various audio/video types are obtained, the ratio of the watched time length of each type of audio/video to the sum of the watched time lengths of all types of audio/video can be determined, and the ratio is used as a pre-loading playing ratio corresponding to the audio/video types, so that the corresponding relation is obtained. The preset duration model is used for determining the watched duration of each type of audio/video. The preset duration model can be obtained by training according to the duration consumed by watching different types of audio/video in the audio/video historical watching records of the target user.

Therefore, the preference of the target user to the target audio/video is determined through the watched time length of each type of audio/video watched by the target user, different pre-loading amounts are configured for different types of audio/video, and the method is easy to realize.

It can be understood that other means may also be adopted to obtain the preference of the target user for the target audio/video, and further obtain the pre-loading amount of the target audio/video, for example, the user information is subjected to big data analysis to obtain a pre-estimation model, and the pre-estimation model is used for determining the pre-loading amount of the target audio/video according to the preference of the target user for the target audio/video, which is not repeated herein.

In one example, the method further comprises:

carrying out Hash operation on the URL address of the preloaded target audio/video to obtain a key code value, recording the key code value, and deleting the corresponding key code value in the record when the preloaded audio/video is played;

when receiving an instruction for playing the target audio/video, carrying out hash operation on the URL address of the target audio/video to obtain a key code value;

judging whether the target audio/video is preloaded or not by judging whether the obtained key code value exists in the record or not;

Therefore, in the embodiment, when the instruction for playing the target audio/video is received, the hash operation is performed on the URL address of the target audio/video to obtain the key code value, the obtained key code value is used for indexing, whether the target audio/video is preloaded is determined when the key code value exists, whether the target audio/video is not preloaded is determined when the key code value does not exist, whether the target audio/video is preloaded is determined by determining whether the key code value exists, and therefore, the determination efficiency can be improved.

In another example, the method further comprises:

Therefore, according to the embodiment, the corresponding relation between the key value and the audio/video data obtained through preloading is established, the audio/video data obtained through preloading by using the key value index can be obtained, and the index efficiency can be further improved.

The various technical features in the above embodiments can be arbitrarily combined, so long as there is no conflict or contradiction between the combinations of the features, but the combination is limited by the space and is not described one by one, and therefore, any combination of the various technical features in the above embodiments also belongs to the scope disclosed in the present specification.

Corresponding to the embodiment of the audio/video preloading method, the disclosure also provides an audio/video preloading device, a device applied by the device and an embodiment of a storage medium.

As shown in fig. 4, fig. 4 is a block diagram of an audio/video preloading device shown in accordance with an exemplary embodiment of the present disclosure, the device comprising: a preload amount determination module 41 and a preload operation module 42.

Wherein the pre-loading amount determining module 41 is configured to determine the pre-loading amount of the target audio/video based on the preference of the target user for the target audio/video.

And the preloading operation module 42 is configured to preload the target audio/video according to the determined amount of preloading, and preload the obtained audio/video data for being played when receiving an instruction to play the target audio/video.

As shown in fig. 5, fig. 5 is a block diagram of another audio/video preloading device shown in the present disclosure according to an exemplary embodiment, on the basis of the foregoing embodiment shown in fig. 4, the pre-loading amount determining module 41 includes:

a probability determination sub-module 411 configured to predict a click probability of the target user clicking the target audio/video based on user information of the target user, the user information including one or more of a profile, a historical viewing record of audio/video.

A pre-load determination sub-module 412 configured to determine a pre-load of the target audio/video according to the click probability.

In an alternative implementation, the target user's profile includes one or more of the target user's gender, age, occupation, location domain; the audio/video historical watching records comprise one or more of watched time length, click times, watching integrity and belonged classification of the audio/video.

As shown in fig. 6, fig. 6 is a block diagram of another audio/video preloading device shown in the present disclosure according to an exemplary embodiment, which is based on the foregoing embodiment shown in fig. 5, and uses each item of information in the user information as a feature value of an interest feature, where the user information corresponding to each audio/video constitutes a feature group, and the feature group includes a feature value set of the interest feature; the apparatus further comprises a weight value determining module 43 configured to: aiming at a feature group corresponding to audio/video, taking a feature value of an interest feature in the feature group as input data, and taking whether the video is clicked or not as output data to construct a training sample set; and training the initialization preference model by using the constructed training sample set to obtain the weight value of each interest feature.

The probability determination submodule 411 is specifically configured to: acquiring a feature group corresponding to a target audio/video; and obtaining the click probability of the target audio/video clicked by the target user based on the obtained feature group and the weight value of the interest feature.

As shown in fig. 7, fig. 7 is a block diagram of another audio/video preloading device shown in the present disclosure according to an exemplary embodiment, based on the foregoing embodiment shown in fig. 5, the pre-loading amount determining sub-module 412 includes:

the factor obtaining sub-module 4121 is configured to obtain an influence factor of a conversion relationship, where the conversion relationship is a conversion relationship between a click probability and a pre-loading amount, the influence factor includes one or more of a network type to which a current network belongs and an available storage amount of a buffer, and the network type includes a WiFi network and a mobile data network.

An information conversion sub-module 4122 configured to convert the click probability into a pre-load amount of the target audio/video based on a conversion relationship corresponding to the influence factor.

In an optional implementation manner, the conversion relationship is a conversion ratio, the influence factor includes a network type to which a current network belongs, and the conversion ratio corresponding to the WiFi network is greater than the conversion ratio corresponding to the mobile data network.

In an optional implementation manner, the impact factor includes an available storage amount of a buffer, and a conversion relationship corresponding to the impact factor is: and the available memory space belongs to a conversion ratio corresponding to the capacity range, and the capacity range and the conversion ratio are in positive correlation.

In an optional implementation manner, the pre-loading amount determining module 41 is specifically configured to:

determining the audio/video type of the target audio/video;

the preload condition comprises one or more of the following conditions:

monitoring the starting of an application program for playing the audio/video;

monitoring a scroll stop event of an audio/video list;

In an alternative implementation manner, the target audio/video is the audio/video currently shown in the screen in the audio/video list, or the target audio/video is the audio/video currently shown in the screen in the audio/video list and the audio/video to be shown in the screen after the scrolling operation is performed according to the page scrolling direction prediction.

As shown in fig. 8, fig. 8 is a block diagram of another audio/video preloading device shown in the present disclosure according to an exemplary embodiment, which is based on any one of the foregoing embodiments shown in fig. 4 to 7, and further includes:

and the hash operation module 44 is configured to perform hash operation on the URL address of the preloaded target audio/video to obtain a key value.

A relationship establishing module 45 configured to establish a corresponding relationship between the key value and the audio/video data obtained by preloading.

And the information judging module 46 is configured to, when the instruction for playing the target audio/video is received, perform indexing according to the key code value corresponding to the target audio/video to judge whether the target audio/video is preloaded.

And an audio/video playing module 47 configured to play the audio/video data obtained by preloading if the target audio/video has been preloaded.

Fig. 8 is illustrated on the basis of fig. 4.

Correspondingly, the present disclosure also provides an electronic device, which includes a processor; a memory for storing processor-executable instructions; wherein the processor is configured to:

Accordingly, the present disclosure also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods described above.

The present disclosure may take the form of a computer program product embodied on one or more storage media including, but not limited to, disk storage, CD-ROM, optical storage, and the like, having program code embodied therein. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage may be implemented by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to: phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic tape storage or other magnetic storage devices, or any other non-transmission medium, may be used to store information that may be accessed by a computing device.

The specific details of the implementation process of the functions and actions of each module in the device are referred to the implementation process of the corresponding step in the method, and are not described herein again.

For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. One of ordinary skill in the art can understand and implement it without inventive effort.

As shown in fig. 9, fig. 9 is a block diagram of an apparatus for audio/video preloading according to an exemplary embodiment of the present disclosure. The apparatus 900 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc. terminal with audio/video playing function.

Referring to fig. 9, apparatus 900 may include one or more of the following components: processing component 902, memory 904, power component 906, multimedia component 908, audio component 910, input/output (I/O) interface 912, sensor component 914, and communication component 916.

The processing component 902 generally controls overall operation of the device 900, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. Processing component 902 may include one or more processors 920 to execute instructions to perform all or a portion of the steps of the methods described above. Further, processing component 902 can include one or more modules that facilitate interaction between processing component 902 and other components. For example, the processing component 902 can include a multimedia module to facilitate interaction between the multimedia component 908 and the processing component 902.

The memory 904 is configured to store various types of data to support operation at the apparatus 900. Examples of such data include instructions for any application or method operating on device 900, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 904 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The power supply component 906 provides power to the various components of the device 900. The power components 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 900.

The multimedia component 908 comprises a screen providing an output interface between the device 900 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 908 includes a front facing camera and/or a rear facing camera. The front camera and/or the rear camera may receive external multimedia data when the device 900 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 910 is configured to output and/or input audio signals. For example, audio component 910 includes a Microphone (MIC) configured to receive external audio signals when apparatus 900 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 904 or transmitted via the communication component 916. In some embodiments, audio component 910 also includes a speaker for outputting audio signals.

I/O interface 912 provides an interface between processing component 902 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor component 914 includes one or more sensors for providing status assessment of various aspects of the apparatus 900. For example, sensor assembly 914 may detect an open/closed state of device 900, the relative positioning of components, such as a display and keypad of device 900, the change in position of device 900 or one of the components of device 900, the presence or absence of user contact with device 900, the orientation or acceleration/deceleration of device 900, and the change in temperature of device 900. The sensor assembly 914 may include a proximity sensor configured to detect the presence of a nearby object in the absence of any physical contact. The sensor assembly 914 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 914 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 916 is configured to facilitate communications between the apparatus 900 and other devices in a wired or wireless manner. The apparatus 900 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 916 receives a broadcast signal or broadcast associated information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 916 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the apparatus 900 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

In an exemplary embodiment, a non-transitory computer readable storage medium comprising instructions, such as the memory 904 comprising instructions, executable by the processor 920 of the apparatus 900 to perform the above-described method is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

Wherein the instructions in the storage medium, when executed by the processor, enable the apparatus 900 to perform an audio/video preloading method comprising:

Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.

It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

The above description is only exemplary of the present disclosure and should not be taken as limiting the disclosure, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present disclosure should be included in the scope of the present disclosure.

Claims

1. An audio/video preloading method, the method comprising:

preloading the target audio/video according to the determined preloading amount, wherein the audio/video data obtained by preloading is used for being played when receiving an instruction for playing the target audio/video;

each item of information in the user information of the target user is taken as a characteristic value of an interest characteristic, the user information corresponding to each audio/video forms a characteristic group, and the characteristic group comprises a characteristic value set of the interest characteristic;

the determining the pre-loading amount of the target audio/video based on the preference of the target user for the target audio/video comprises the following steps:

acquiring a feature group corresponding to a target audio/video;

obtaining the click probability of the target audio/video clicked by the target user based on the obtained feature group and the weight value of the interest feature; the weight value of the interest feature indicates the influence degree of the interest feature on the audio/video preference of the target user;

2. The method of claim 1, wherein determining the amount of pre-loading of the target audio/video based on the target audio/video preference of the target user further comprises:

3. The method of claim 2, wherein the user information includes one or more of a profile, a historical viewing record of audio/video; the personal data of the target user comprises one or more of gender, age, occupation and location domain of the target user; the audio/video historical watching records comprise one or more of watched time length, click times, watching integrity and belonged classification of the audio/video.

4. The method of claim 2, further comprising:

and training the initialization preference model by using the constructed training sample set to obtain the weight value of each interest feature.

5. The method of claim 2, wherein said determining a pre-load amount of said target audio/video according to said click probability comprises:

6. The method of claim 5, wherein the transition relationship is a transition ratio, the impact factor includes a type of a network to which a current network belongs, and the transition ratio corresponding to the WiFi network is greater than the transition ratio corresponding to the mobile data network;

7. The method of claim 1, wherein determining the amount of pre-loading of the target audio/video based on the target audio/video preference of the target user comprises:

determining the audio/video type of the target audio/video;

8. The method of any of claims 1 to 7, wherein determining the amount of pre-loading of the target audio/video based on the target audio/video preference of the target user comprises:

the preload condition comprises one or more of the following conditions:

monitoring the starting of an application program for playing the audio/video;

monitoring a scroll stop event of an audio/video list;

9. The method according to claim 8, wherein the target audio/video is an audio/video currently displayed on a screen in an audio/video list, or the target audio/video is an audio/video currently displayed on a screen in an audio/video list and an audio/video to be displayed on a screen after performing a scrolling operation according to a page scrolling direction prediction.

10. The method according to any one of claims 1 to 7, further comprising:

11. An audio/video preloading device, the device comprising:

the preloading operation module is configured to preload the target audio/video according to the determined preloading amount, and the audio/video data obtained by preloading is used for being played when the target audio/video playing instruction is received;

the pre-loading capacity determining module comprises a probability determining submodule and a pre-loading capacity determining submodule;

the probability determination submodule is configured to obtain a feature group corresponding to a target audio/video; obtaining the click probability of the target audio/video clicked by the target user based on the obtained feature group and the weight value of the interest feature; the weight value of the interest feature indicates the influence degree of the interest feature on the audio/video preference of the target user;

the pre-load determination sub-module is configured to determine a pre-load of the target audio/video according to the click probability.

12. The apparatus of claim 11, wherein the pre-loading determination module comprises:

13. The apparatus of claim 12, wherein the user information comprises one or more of a profile, a historical viewing record of audio/video; the personal data of the target user comprises one or more of gender, age, occupation and location domain of the target user; the audio/video historical watching records comprise one or more of watched time length, click times, watching integrity and belonged classification of the audio/video.

14. The apparatus of claim 12,

the apparatus further comprises a weight value determination module configured to:

15. The apparatus of claim 12, wherein the pre-loading determination sub-module comprises:

16. The apparatus of claim 15, wherein the transition relationship is a transition ratio, the impact factor includes a type of a network to which a current network belongs, and the transition ratio corresponding to the WiFi network is greater than the transition ratio corresponding to the mobile data network;

17. The apparatus of claim 11, wherein the pre-loading determination module is specifically configured to:

determining the audio/video type of the target audio/video;

18. The apparatus of any of claims 11 to 17, wherein the pre-loading determination module is specifically configured to:

the preload condition comprises one or more of the following conditions:

monitoring the starting of an application program for playing the audio/video;

monitoring a scroll stop event of an audio/video list;

19. The apparatus according to claim 18, wherein the target audio/video is an audio/video currently displayed on the screen in the audio/video list, or the target audio/video is an audio/video currently displayed on the screen in the audio/video list and an audio/video to be displayed on the screen after the scrolling operation is performed according to the page scrolling direction prediction.

20. The apparatus of any one of claims 11 to 17, further comprising:

21. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor is configured to:

acquiring a feature group corresponding to a target audio/video;

22. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 10.